Silver Salt Analysis

This document describes how to set up and run nitrate or sulfate for Δ¹⁷O on the Lorax, ThermoFinnigan MAT 253. Silver nitrate samples (loaded into silver capsules) or silver sulfate samples (loaded into quartz or gold capsules) are dropped by an autosampler into a hot quartz pyrolysis reactor within a Temperature Conversion Elemental Analyzer (TCEA). A helium carrier stream moves pyrolysis products through a liquid nitrogen trap where byproducts other than O₂ are frozen. The O₂ product then continues through a gas chromatograph and either a Conflo or Gasbench interface. The Conflo is used for samples of O₂ &grt; 2 μmol; the Gasbench cryofocuses O₂, enabling analysis of smaller samples (50-1000 nmol). Make sure the sample is <1000nmol when using Gasbench mode. Helium continues to move O₂ into the mass spectrometer for m/z 32, 33, and 34 measurements.

For clarity, sections specific to the Conflo mode will have a Conflo tag while the Gasbench mode will have a Gasbench tag.

This general plumbing diagram and this gasbench valco valve diagram may help your understanding of the system.

Risk associated with this method include extreme temperatures.

Appropriate precautions should be taken to protect yourself against very cold temperatures as you work with liquid nitrogen. Wear appropriate gloves and eye protection to protect against cryogenic burns.

Very hot temperatures will be encountered as you work with the TCEA and its associated parts. Wear appropriate hand and eye protection when working with the hot reaction column or baking out trap to prevent burns.

Nomenclature - All of the terms, acronyms, and colloquialisms that apply to this SOP can be found in the nomenclature section below.

Conflo vs Gasbench - Samples can be analyzed across a reasonably wide range of sulfate or nitrate amounts. The larger samples (e.g. 1-10 µmol) are measured using Conflo mode while smaller samples (e.g. <1 µmol) are measured using Gasbench mode. Switching between these two modes requires careful attention to detail and patience.

Pyrolysis column - The pyrolysis column is the temperature-controlled location of the reaction inside the TCEA. Silver nitrate samples use a reaction temperature of 585 °C and silver sulfate samples use a reaction temperature of 1000 °C. The pyrolysis column needs to be cleaned out after each run of samples.

Sample Loading - Most of the work associated with loading samples into the instrument is spent purging atmospheric air out of the autosampler. Take your time in this section to minimize backgrounds and the size of the O₂ blank.

Run sequence - You will create a run sequence in the Isodat software that will include sample identifiers, amount of material, run method, and other details pertaining to each individual sample capsule.

Daily Log - We use daily logs for all our instrumentation to keep us organized, account for usage, and make certain the instrument is ready for our samples.

Start your run - If you have made it this far, the instrument must be ready and it is time to start your run.

Instrument bake-out - Consider baking out the instrument after each run. This may not be required but baking the entire set-up is at least an overnight time commitment.

Troubleshooting - If something does not seem right, check here.

Suggested Reading - There are always more papers to read.

Peripherals

• TCEA - The Temperature Conversion Elemental Analyzer (TCEA, or TC/EA) is a device that can heat a reactor space up to 1400°C depending on the reaction of interest. We use 585°C for nitrate conversion to O₂ plus byproducts and 1000°C for sulfate conversion to O₂ plus byproducts.
• Conflo - The Conflo is short for continuous flow and is a device that allows a high flow rate peripheral (e.g. TCEA at ~80 mL / min) to interface with a low flow rate mass spectrometer (~2 mL / min). This is achieved by venting ~95% of your sample out the vent and up through the hood (i.e. most of your sample is lost). The conflo is used when you are not sample limited.
• Gasbench - The Gasbench is a general purpose device for those that do not wish to lose most of their sample as happens with the Conflo. High flow rate carrier gas with an entrained sample gas flows through a cryogenic trap, collecting the sample (O₂ in our case). Once collection is complete, the gasbench is switched to a low flow state and transfers the sample O₂ to another cryogenic trap for subsequent separation in a gas chromatograph (GC) and then analysis in the mass spectrometer.
• Dual-inlet - While we do not use the dual-inlet for silver salt pyrolysis, Lorax does have one and you may need to consider it from time to time. You may need to switch from dual-inlet to conflo or gasbench at some point. The dual-inlet is a pressure and vacuum based measurement for high precision analyses and it requires a large amount of O₂. Our lab uses this method to measure the Δ¹⁷O of O₂ from water and carbonate. This method does not use a helium carrier gas like the conflo and gasbench.

Liquid nitrogen traps

• TCEA pre-cleanup trap - This is a ¼” loop of stainless steel tubing filled with 5A molecular sieve submerged in liquid nitrogen to remove O₂ from the helium carrier before entering the reactor.
• TCEA post-cleanup trap - This is a 1/16” loop of stainless steel tubing submerged in liquid nitrogen to remove unwanted post-pyrolysis byproducts (e.g. NO_x and SO₂).
• Big trap or Concentrating trap - This is a 1/16” loop of stainless steel tubing packed with 5A molecular sieve as part of the Gasbench. This trap will collect all of the sample O₂ that came from the TCEA. The Isodat software refers to this as "Trap 2".
• Little trap or Cryofocusing trap - This is a fused silica capillary loop packed with 5A molecular sieve as part of the Gasbench. In principle, this trap focuses the sample O₂ into a tighter bolus to yield a narrower peak. The Isodat software refers to this as "Trap 1".

Valves

• Autosampler Purge Valve - This is a three-way valve on top of autosampler carousel used while removing atmospheric air from the autosampler after loading samples. The valve position dictates evacuation of the autosampler, venting of helium with atmospheric air still mixed with it, or closure.
• He Purge Valve - This is a sliding-barrel valve located on the left side of the autosampler carousel and is used when helium is needed separately for the purposes of purging the carousel and autosampler as described above. It would only be used if the Isolation Valve (see below) is closed.
• Isolation Valve - This is the large wheel or knob located between the autosampler carousel and the reactor column to isolate the autosampler from the reactor when loading samples.
• Vent Valve - This is another three-way valve located on the rear and top of the TCEA and is used to vent the carrier gas after installing a fresh pyrolysis column or while initially heating the clean-up trap ("Vent" position) or to allow the helium carrier stream to continue to the TCEA GC or the Gasbench ("Run" position)
• Method Valve - This is a four-way valve located on the rear and top of the TCEA to choose between Gasbench or Conflo modes.
• Auxillary He Valve - This is a two-way valve next to Method Valve and is used to allow helium flow to both methods (Gasbench and Conflo modes). It would be used only if you are interested in running both modes, albeit separately, and want to keep each system clean and purged with helium.
• Valco Valve - This is an eight-way valve located inside the Gasbench and is only used when in Gasbench mode. The Isodat software toggles this valve to "Load" or "Inject" automatically during a run. You can also toggle the Valco valve by clicking on it in the Gas Bench window of the Isodat software. This valve is not used in Conflo mode.
• Needle Valves - The needle valves are used to switch between Gasbench, conflo, or dual-inlet. Only one is open at a time or both are closed. The needle valves are located at the mass spectrometer inlet on the west side of the flight tube above the large red cable. The top valve is from the Gasbench and the side valve is from the Conflo.

Samples can be analyzed across a reasonably wide range of sulfate or nitrate amounts. The larger samples (e.g. 1-10 µmol) are measured using Conflo mode while smaller samples (e.g. <1 µmol) are measured using the Gasbench mode. The Gasbench is further separated into two modes: "Split-out" and "Split-in". Split-out is for 300 to 1100 nmol samples while Split-in should be used for tiny samples (e.g. <300 nmol). These two modes are named after the peripheral that precedes the inlet to the mass spectrometer.

Adjusting the needle valves

If you are switching from one mode to another, you will need to close one needle valve and open the other. You may also be switching from dual-inlet mode to conflo or gasbench, in which case you will need to open the appropriate needle valve. Use the photo above to identify the specific needle valve of interest.

1. To close a needle valve, make sure the vacuum is visible from within the Isodat software (you can hover the mouse over the high vacuum reading and it will magnify) while you stand at the needle valves. Close the valve ONLY as much as needed. YOU CAN CLOSE IT TOO MUCH AND RUIN THE VALVE. Watch the vacuum reading and begin twisting the needle valve closed (clockwise, righty tighty). When vacuum decreases to <3e-8 mbar, stop twisting. WAIT. If vacuum doesn’t appear to be decreasing any more, GENTLY close the needle valve some more. If the vacuum responds, WAIT and repeat.
2. The appropriate needle valve should be open the day before you intend to run to allow the fused silica capillary time to purge with helium.
3. Conflo - Make certain the dilution is ON (within the left side of the Isodat software under the Conflo panel) before attempting to open the needle valve.
4. Just as with closing a needle valve described above, make the vacuum visible while you are standing at the needle valves.
5. VERY SLOOOOOWWWWLY open the appropriate needle valve watching the vacuum. You will see the vacuum rise from <2e-8 mbar to ~1e-6 mbar (it may be in the 9e-7 range). If the vacuum reading jumps into the Xe-5 range and stays there, close the needle valve. Verify that helium is in fact flowing to the peripheral of interest and try again. If it still remains in the Xe-5 range, close the needle valve and seek advice.
6. If you have just opened a needle valve that was previously closed, run an 8 hour bake-out sequence for the appropriate mode (Conflo vs Gasbench).

Software

In the lower left corner of Isodat, use the menu to select Conflo or GasBench.

The pyrolysis column is an empty quartz reactor with a restriction at the base of the TCEA hot zone (the vertical space along which the temperature is held at your set point) to catch sample capsules and allow the reaction to take place at the set temperature. The sample capsules must be cleaned out after each run. The reactor can accommodate many more nitrate samples than sulfate samples because the nitrate samples are wrapped up tightly into a small silver capsule while sulfate samples are loaded into more bulky quartz or gold capsules. About 10 quartz capsules will fit into the reactor before they start piling up high enough to leave the hot zone. Silver metal and other condensables adhere to the side walls outside of the hot zone and eventually may affect your analysis (see anti-blank below) making it best to start with a new reactor.

As of today (2022-09-27), the column design has a pinch point at 32 cm from the top of the column. We are using quartz tubes from EA suppliers that have a 2 mm wall thickness. The pinch is created on a vacuum line in the wet lab and the "quartz" torch tip. The hot-zone based on the discoloration of the column after many sulfate samples (and as of 2022-09-27) is between 22 cm and 32.5 cm.

1. You will either need to clean out the previous runs capsules or replace the pyrolysis tube before starting a new run.
2. Conflo - Turn the He dilution ON if running Conflo mode.
3. Gasbench - Change the Valco Valve to Inject if running in Gas Bench mode.
4. Turn the Vent Valve to vent.
5. Turn the Autosampler 3-way valve to vent.
6. Use gloves and protective eye-wear. Unscrew the pyrolysis column starting at the top by loosening the nut all the way, removing the autosampler, and setting the autosampler on top of and towards the back of the TCEA.
7. Unscrew the bottom fitting and remove the body, nut and two teflon ferrule pieces.
8. Remove pyrolysis tube by pushing the bottom up and grabbing the metal bushing and o-ring and place in the pyrex baking dish.
9. Select a new pyrolysis tube or clean out the existing one with the wire-tipped wand.
10. Place the metal bushing and rubber o-ring on top of the new pyrolysis column about 5 mm from the top. Use compressed air (red tube above TCEA) to clean any dust off the teflon ferrule. Slide pyrolysis tube into TCEA. Start at the top of the TCEA. While bearing the weight of the autosampler, screw on the top column nut and use a spanner to tighten an additional quarter turn. Now move to the bottom of the column. Use compressed air to cool the bottom of the tube that just went through the furnace. Slide fitting onto the pyrolysis tube as high as it can go and ensure you have the proper order of the nut, back of ferrule, and front of ferrule and then finger tighten. Use wrenches to tighten an additional quarter to half turn. DO NOT OVERTIGHTEN.
11. The top rubber ferrule will compress over time, but will recover if allowed to rest for a few days – swap it with another rubber ferrule when it looks compressed or is leaking. The bottom Teflon ferrules compress over multiple uses. Once they start leaking (or get dirty), they must be replaced with new ferrules.
12. Close the 3-way valve.
13. Use the leak detector to check for leaks. Check around both top and bottom fittings. If leaks are present, consider tightening fittings or repeat process of loading the pyrolysis tube without actually removing the column.
14. Once leaks are eliminated, leave Vent Valve in vent position for ~10 minutes before switching to run.

Each sulfate run may contain a maximum of 10 samples in quartz or gold capsules. Nitrate runs may contain up to a full carousel (49 balled silver capsules). The goal in this section is the load your samples and then ensure all of the atmospheric air that was allowed into the autosampler is purged out and replaced with helium.

1. Isolate the autosampler by closing the Isolation Valve. Rotate the valve 90°; the arrow on side of valve will be pointing towards you and you won’t be able to see down inside the reactor.
2. Turn the Autosampler 3-way Valve clockwise to vent pressure from inside the autosampler. The valve handle will point towards the Lorax computer.
3. Loosen the three knurled nuts that clamp the autosampler lid into place (you may need a spanner). Open the lid and use compressed air to clean out any debris.
4. Load samples into slots. Close the lid, and gradually hand tighten each clamp a little at a time, alternating between nuts to ensure even pressure.
5. Turn Autosampler 3-way Valve 180° counter-clockwise to evacuate the autosampler. While it’s evacuating, tighten the nuts onto the lid with your fingers again.
6. After 3-5 minutes, turn Autosampler 3-way Valve to point straight back (off position).
7. Slowly open He Purge Valve. Monitor pressure on the front of TCEA until it levels off (Purge=1.5 bars and Carrier=1.5 bars).
8. Use leak detector to check for autosampler leaks around the lid as was done for the reactor.
9. If no leaks are found, slowly turn the Autosampler 3-way Valve clockwise until the helium vents and the pressure gauges drop. Then turn the Autosampler 3-way Valve counter clockwise to the closed (straight back) position. Once the pressure gauges read 1.5 bars, vent the helium again and repeat 10-12 times.
10. Conflo - Ensure the dilution is ON.
11. Gasbench - ensure the Valco Valve is in the Inject position.
12. Open the Isolation Valve (rotate it until the arrow is pointing up).
13. Conflo - Monitor the O₂ background and keep venting helium from the autosampler as in the above step until the backgrounds are sufficiently low.
14. Gasbench - We can't use real-time backgrounds because the TCEA is never in a direct flow path with the mass spectrometer and we must use a different technique outlined below. For now, we will blindly do our best.
15. Conflo - Monitor the O₂ background signal for ~5 minutes. If the background has not increased dramatically since opening the Isolation Valve, turn the dilution off.
16. Place the clean-up traps in the dewar and fill the dewar with liquid nitrogen.
17. Close the purge valve.
18. Vent the helium using the Autosampler 3-way Valve as above but this time the pressure gauges will NOT drop (because the purge valve is closed). Wait about one minute. Repeat this venting 5-10 times.
19. Conflo - Monitor the O₂ background signal for ~ 5 minutes. If it is not rising proceed to the next section, otherwise repeat the above venting procedure and or check for leaks.
20. Gasbench - Proceed to the next section but realize you may purge more after you see the size of your blanks.

Create a sequence using the IsoDat Acquisition software (if Acquisition is already running something you can use Workspace instead). All of the data you type into the sequence are exported with the data. Isodat saves the sequence file upon starting a run. Saving a unique version of the sequence itself is not required.

1. Conflo - Generally use the Conflo_O2_Samples.seq sequence.
2. Gasbench - Generally, use the GasBench_O2_Samples.seq.
3. Make sure a check is in every box in the "Peak Center" column.
4. Record the amount of expected O₂ for each sample (μmol or nmol) under "Amount".
5. Enter sample name or ID under "Identifier 1" and any other information under "Identifier 2", "Comment", and "Preparation"
6. Make sure the appropriate method (*.met) is selected under the "Method" header.
   • Conflo - We usually have a “Zero” at the beginning and end of the run, as well as in between each sample. Make sure that Zero runs use method Conflo_O2_Zero.met and sample runs use method Conflo_O2_Sample.met.
   • Gasbench - Include at least three “Blanks” at the beginning and end of the run, as well as at least one in between each sample. Make sure the blanks use method GasBench_O2_Blk_*.met and samples use GasBench_O2_Sample_*.met, where * is either "SplitIN" or "SplitOUT". We suggest using "SplitOUT" for samples amount between 300 and 1100 nmol.
7. Save the sequence. Again, saving a unique version of the sequence itself is not required because everything you enter into a sequence is written to the exported data file.

Check List

Well done. You made it this far. We will go through a quick check list and then start.

• The Autosampler 3-way Valve is closed.
• The Isolation Valve is open.
• The Helium Purge Valve is closed.
• The Mode Valve is set according to the mode you wish to be in: Conflo vs Gasbench.
• The appropriate Needle Valve is open and the other Needle valve is closed.
• The TCEA Clean-up traps are submerged in liquid nitrogen.
• Gasbench - The large blue dewar is full of liquid nitrogen and placed under the Gasbench traps.
• The liquid nitrogen dewar(s) will need to be topped off several times during the course of a day, so check periodically to ensure appropriate levels.

Start the run

• In IsoDat Acquisition and in with your sequence, if you are not running every row in the sequence, highlight the appropriate rows by clicking on the "Line" column and dragging down.
• Click the green “Start” button at the top of the window.
• Fill in the "Folder Name" field with your SampleSetID and perhaps run number. For example "Lorax_210611_Run1". Clicking on the "Pre" button should show you that "Data" is checked.
• Leave the "File Name" field blank because the "Pre" has "Analysis" checked and "Post" has "Identifier 1" checked. Analysis is a unique serial number assigned to everything run through the instrument. Identifier 1 is what you typed into the sequence.
• Copy and paste or type the same thing into "Export File Name" that you entered above into "Folder Name".
• Under "Modify Template List" should be either "O2_Gasbench.wke" or "O2_Conflo.wke".
• The format should be "ASCII (*.csv)" but if you click on the "Modify Template List" button, the format will default to "Excel (*.xls)" and you will need to change it back.
• Click ‘OK’ and your run will start.

Now What?

• Conflo - Wait for at least one sample peak to come off so that you know things are working.
• Gasbench - Look carefully at the size of the blanks. If they are too large (e.g. > 5 Vs), consider purging the autosampler more and starting the run over.
• If everything seems in order, take a break and come back at some point to top of the dewars with more liquid nitrogen.
• When the run is finished, proceed to the bake-out section below.

For maximum efficiency, these steps happen immediately after a run has finished. It is also good practice to complete the bake-out routine before (not necessarily immediately before) starting another set of samples. If you are still in the lab after a run finishes and have about 30 minutes, complete these steps. Otherwise, try to complete them the following day or at least prior to a future run.

Liquid nitrogen and pyrolysis column

1. When a run is done (in either Conflo or Gasbench mode), turn the vent valve to “Vent” mode and thaw the TCEA cleanup trap by removing it from the liquid nitrogen dewar and heating it on ‘HI’ for 5 minutes. This is designed to remove most of the trapped SO₂ that may be present and allow it to go out the vent through the snorkel hood in the ceiling.
2. Pour any leftover liquid nitrogen back into a storage dewar.
3. Clean out the pyrolysis column as described above under Pyrolysis Column.
4. Change the Vent Valve back to run mode.
5. When running is Gas Bench mode, make sure the Valco valve is in “Inject” mode (toggle on the IsoDat control screen). Pour any leftover liquid nitrogen from the large blue dewar back into a storage dewar.

8-hour Gasbench Bake Routine - Use this bake sequence when actively running in Gasbench mode or if you have just switched to Gasbench mode and want to give the system and initial bake.

• Helium Clean-up Trap:
  1. This sequence bakes out the two clean-up traps that are wrapped in heat tape to the left of the TCEA.
  2. Face the TCEA and remove the liquid nitrogen dewar that is on the clean-up traps
  3. set the "TCEA Helium clean-up trap" controller to "HI"
• Gasbench Trap Heater Box:
  1. This sequence bakes out the two Gasbench traps that continuously go up and down over the course of a run.
  2. Face the Gasbench and remove the large blue liquid nitrogen dewar
  3. remove the left-side panel
  4. hold the heater box up in place while placing the lab jack underneath it
  5. slide the two covers over the top of the box
  6. attach the left-side panel back in place
  7. Face the TCEA and set the "Gasbench Trap Heater Box" controller to 160 °C (#3 on the dial)
• Gasbench GC:
  1. This sequency starts a 7 hour bake of the Gasbench GC whereupon the first hour is heating up to 200 °C, it bakes for about 6 hours at 200 °C, and then turns off the heater and cools down for the last hour.
  2. Face the side of the Gasbench with the Jumo iTron 16 controller
  3. Press P until you see "SP"
  4. Arrow up until value reads "200" (you can hold down the up arrow)
  5. Press P again and you will see "t_'0"
  6. Press the up arrow once
• Isodat software:
  1. This sequence starts the Isodat mass spec software on an 8 hour bake routine.
  2. On the computer, in the left panel under the "File Browser" and open the "GasBench_O2_Bake_8hr.seq" sequence.
  3. Without any rows of the three-row sequence highlighted, click the "Start" button

8-hour Conflo Bake Routine - Use this bake sequence when actively running in Conflo mode or if you have just switched to Conflo mode and want to give the system and initial bake.

• Helium Clean-up Trap:
  1. This sequence bakes out the two clean-up traps that are wrapped in heat tape to the left of the TCEA.
  2. Face the TCEA and remove the liquid nitrogen dewar that is on the clean-up traps
  3. set the "TCEA Helium clean-up trap" controller to "HI"
• TCEA GC:
  1. This sequency starts a 7 hour bake of the TCEA GC whereupon the first hour is heating up to 200 °C, it bakes for about 6 hours at 200 °C, and then turns off the heater and cools down for the last hour.
  2. Face the TCEA and find the Jumo iTron 16 controller for the "GC"
  3. Press P until you see "SP"
  4. Arrow up until value reads "200" (you can hold down the up arrow)
  5. Press P again and you will see "t_'0"
  6. Press the up arrow once
• Isodat software:
  1. This sequence starts the Isodat mass spec software on an 8 hour bake routine.
  2. On the computer, in the left panel under the "File Browser" and open the "Conflo_O2_Bake_8hr.seq" sequence.
  3. Without any rows of the three-row sequence highlighted, click the "Start" button

Ending the Bake Routine

• Turn off the helium clean-up trap heater.
• Conflo - Turn the TCEA GC down to 0 (or at least below 20 °C) if you did not use the automated timer feature of the Jumo iTron controller.
• Gasbench - Turn the Gasbench GC down to 0 (or at least below 20 °C) if you did not use the automated timer feature of the Jumo iTron controller.
• Gasbench - Turn the Gasbench trap heater box off, put on leather gloves, and remove the heater box.

While the data processing is entirely up to you, I suggest running the python script loraxGB.py and then loraxGBcalibrate.py. The goal here is the monitor instrument performance and be able to easily compare new runs with historical runs.

1. Copy the .csv file from the Lorax computer to the data server (S:). The python scripts are setup to use S:\Data\lorax\gasbench\raw\.
2. On one of the Zax computers, open a cmd prompt, type 'S:', then enter, then 'cd Data\python\', then enter.
3. Type 'py loraxGB.py' and hit enter. This takes your recent run(s) and puts all the data into S:\Data\projects\lorax\gasbench\loraxGB_analysis_log.csv, which is a log file of all analyses for the current session.
4. Type 'py loraxGBcalibrate.py'. This reads in the log file mentioned above and creates figures saved to an .html page.
5. In S:\Data\projects\lorax\gasbench\figs\, double click on 'open-me-to-view-figures.html' and the latest figures will be viewable in the defualt browser.
6. That is it for now. Certainly more needs to be done here...

• Background levels are high - The autosampler may not be fully purged with helium since loading new capsules. Try re-purging the autosampler a few times. Another option is that the column is not leak free at either the top or bottom connection. Check with the helium leak detector.
• Background levels are still high - If you know nothing is leaking and the autosampler is purged properly, the maybe the GC/Cryo Traps are retaining remnants of atmospheric air or other mass interfering species. If you did not bake everything out recently, try completing the bake out procedure above. This will take at least overnight.
• The sample peak is delayed in Conflo mode - This is very likely a leak somewhere (probably the reactor). Check top and bottom fittings for tightness and use the helium leak detector.
• The sample peak shape off or sample peak timing early when in Gasbench mode - The molecular sieve cryo traps may be wet or are otherwise contaminated. If you did not bake everything out recently, try completing the bake out procedure above. This will take at least overnight.
• The sample peak is smaller than anticipated when in Gasbench mode - If your samples are particularly small, silver adhered to the side-walls of the quartz reactor from previous samples may be absorbing some of your O₂. You can try a new reactor or try cleaning the existing reactor. It may also be that dropping fewer quartz capsules before cleaning them out is required.
• Sample peaks become tiny or are completely absent after being present in a run - Sometimes the quartz columns break when a quartz capsule hits the pinch points which will cause a massive helium leak and prevent sample O₂ from getting to the mass spec. Investigate and install a new pyrolysis column if needed. You can test columns for leaks using the fittings in the bag labeled “Reactor Vacuum Test” in the cabinet next to the Lorax computer. Put the rubber stopper in one end, attach the other and connect the glass tube to a vacuum line (a couple are next door). If the pyrolysis tube pumps down, it is ready to go.

• Geng L, Schauer AJ, Kunasek SA, Sofen ED, Erbland J, Savarino J, Allman DJ, Sletten RS, Alexander B. (2013). Analysis of oxygen-17 excess of nitrate and sulfate at sub-micromole levels using the pyrolysis method. Rapid Communications in Mass Spectrometry 27, 2411-2419. doi: 10.1002/rcm.6703.
• Kunasek SA, Alexander B, Steig EJ, Sofen ED, Jackson TL, Thiemens MH, McConnel JR, Gleason DJ, Amos HM. (2010). Sulfate sources and oxidation chemistry over the past 230 years from sulfur and oxygen isotopes of sulfate in a West Antarctic ice core. Journal of Geophysical Research Atmospheres115. doi: 10.1029/2010JD013846. Highlighted in Eos Vol. 91, No. 49, 7 December 2010 "Research Spotlight".
• Michalski G, Savarino J, Böhlke JK, Thiemens M. (2002). Determination of the total oxygen isotopic composition of nitrate and the calibration of a Δ17O nitrate reference material. Analytical Chemistry. 74, 4989. doi:10.1021/ac0256282.
• Savarino J, Alexander B, Darmohusodo V, Thiemens MH. (2001). Sulfur and oxygen isotope analysis of sulfate at micromole levels using a pyrolysis technique in a continuous flow system. Analytical Chemistry. 73, 4457. doi:10.1021/ac010017f.
• Schauer AJ, Kunasek SA, Sofen ED, Erbland J, Savarino J, Johnson BW, Amos HM, Shaheen R, Abaunza M, Jackson TL, Thiemens MH, Alexander B. (2012). Oxygen isotope exchange with quartz during pyrolysis of silver sulfate and silver nitrate. Rapid Communications in Mass Spectrometry 26. doi: 10.1002/rcm.6332.